{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "30788e1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "### In this script, we load the sentence data with the classifier predictions and certainty scores.\n",
    "### We then calaculate the societal engagement percentage based on different thresholds of certainty scores."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "083a2aab",
   "metadata": {},
   "outputs": [],
   "source": [
    "### ALL IMPORTS\n",
    "\n",
    "import os\n",
    "import pandas as pd\n",
    "import re\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from PIL import Image\n",
    "\n",
    "from statsmodels.api import OLS, add_constant\n",
    "from scipy import stats\n",
    "import numpy as np\n",
    "from statsmodels.stats.diagnostic import het_breuschpagan\n",
    "from statsmodels.stats.outliers_influence import variance_inflation_factor\n",
    "from statsmodels.stats.stattools import durbin_watson\n",
    "from scipy.stats import linregress\n",
    "from linearmodels.panel import PanelOLS\n",
    "from linearmodels.panel import compare"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "67f51c19",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(99501, 29)\n"
     ]
    }
   ],
   "source": [
    "### Load the aggregated data to link to the paper-level data with the FOS thresholds\n",
    "\n",
    "### directories\n",
    "\n",
    "DIR = \"/Replication_SocietalAI\"\n",
    "DATA_DIR = os.path.join(DIR, \"Data\")\n",
    "GRAPHICS_DIR = os.path.join(DIR, \"Graphics\")\n",
    "if not os.path.exists(GRAPHICS_DIR):\n",
    "    os.makedirs(GRAPHICS_DIR)\n",
    "##### Load the main file\n",
    "df = pd.read_csv(os.path.join(DATA_DIR, \"processed_June2025_final_data.csv\"))\n",
    "print(df.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d9c14e97",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the categories column into individual categories\n",
    "categories_split = df['categories'].str.split()\n",
    "\n",
    "# Filter rows to only include articles with the desired categories\n",
    "categories_of_interest = {'cs.cl', 'cs.cv', 'cs.cy', 'cs.ai'}\n",
    "filtered_df = df[\n",
    "    categories_split.apply(lambda x: any(category in categories_of_interest for category in x))\n",
    "]\n",
    "\n",
    "# Create dummy variables for each unique category\n",
    "dummies = pd.get_dummies(categories_split.apply(pd.Series).stack()).groupby(level=0).sum()\n",
    "\n",
    "# Filter dummy variables to include only the categories of interest\n",
    "dummies = dummies[list(categories_of_interest)]\n",
    "\n",
    "# Merge the filtered dummy variables back into the filtered DataFrame\n",
    "filtered_df = pd.concat([filtered_df, dummies], axis=1)\n",
    "\n",
    "df = filtered_df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "00bc7845",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(14355044, 6)\n"
     ]
    }
   ],
   "source": [
    "## Load the sentence-level data with the classifier predictions and certainty scores\n",
    "sentences = pd.read_csv(os.path.join(DATA_DIR, \"sentences_with_predictions_23032025_WithPaperIDS.csv\"))\n",
    "print(sentences.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1a79e1ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "#### create new column with label based on 0.6 and 0.8 probability thresholds\n",
    "sentences['predicted_label_06'] = sentences['predicted_probability'].apply(lambda x: 'Yes' if x > 0.6 else 'No')\n",
    "sentences['predicted_label_08'] = sentences['predicted_probability'].apply(lambda x: 'Yes' if x > 0.8 else 'No')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "429f2f45",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "predicted_label\n",
      "No     0.923957\n",
      "Yes    0.076043\n",
      "Name: proportion, dtype: float64\n",
      "predicted_label_06\n",
      "No     0.933513\n",
      "Yes    0.066487\n",
      "Name: proportion, dtype: float64\n",
      "predicted_label_08\n",
      "No     0.954939\n",
      "Yes    0.045061\n",
      "Name: proportion, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "### examine the distribution of the predicted probabilities between different thresholds\n",
    "print(sentences.predicted_label.value_counts(normalize=True))\n",
    "print(sentences.predicted_label_06.value_counts(normalize=True))\n",
    "print(sentences.predicted_label_08.value_counts(normalize=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1cf1aa59",
   "metadata": {},
   "outputs": [],
   "source": [
    "##### Find the percentages per document with the different thresholds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "48a5a8a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "### for 0.6\n",
    "# Filter only the relevant columns\n",
    "relevant_df = sentences[['paper_id', 'predicted_label_06']]\n",
    "\n",
    "# Group by paper_id and calculate total and societal counts\n",
    "grouped = relevant_df.groupby('paper_id').agg(\n",
    "    total_sentences=('predicted_label_06', 'count'),\n",
    "    societal_sentences=('predicted_label_06', lambda x: (x == 'Yes').sum())\n",
    ")\n",
    "\n",
    "# Calculate percent_societal\n",
    "grouped['percent_societal_06'] = grouped['societal_sentences'] / grouped['total_sentences'] * 100\n",
    "\n",
    "# Reset index for a cleaner DataFrame\n",
    "percent_societal_df_06 = grouped.reset_index()[['paper_id', 'percent_societal_06','total_sentences']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9dacfffe",
   "metadata": {},
   "outputs": [],
   "source": [
    "### for 0.8\n",
    "# Filter only the relevant columns\n",
    "relevant_df = sentences[['paper_id', 'predicted_label_08']]\n",
    "\n",
    "# Group by paper_id and calculate total and societal counts\n",
    "grouped = relevant_df.groupby('paper_id').agg(\n",
    "    total_sentences=('predicted_label_08', 'count'),\n",
    "    societal_sentences=('predicted_label_08', lambda x: (x == 'Yes').sum())\n",
    ")\n",
    "\n",
    "# Calculate percent_societal\n",
    "grouped['percent_societal_08'] = grouped['societal_sentences'] / grouped['total_sentences'] * 100\n",
    "\n",
    "# Reset index for a cleaner DataFrame\n",
    "percent_societal_df_08 = grouped.reset_index()[['paper_id', 'percent_societal_08','total_sentences']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "aec40d51",
   "metadata": {},
   "outputs": [],
   "source": [
    "#### merge the data\n",
    "\n",
    "merged_data = pd.merge(df, percent_societal_df_06[['paper_id','percent_societal_06']], on='paper_id', how='left')\n",
    "merged_data = pd.merge(merged_data, percent_societal_df_08[['paper_id','percent_societal_08']], on='paper_id', how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "ba9b75f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "bar_data = merged_data.copy()\n",
    "\n",
    "# Shorten category labels\n",
    "label_map_display = {\n",
    "    'CS-only': 'CS-only',\n",
    "    'Interdisciplinary (Social Sciences/Humanities)': 'Interd. (SSH)',\n",
    "    'Interdisciplinary (Natural Sciences/Medicine)': 'Interd. (NSM)',\n",
    "    'Fully Interdisciplinary': 'Interd. (Full)'\n",
    "}\n",
    "bar_data['short_label'] = bar_data['interdisciplinary_category'].map(label_map_display)\n",
    "\n",
    "# Group and compute means\n",
    "means = bar_data.groupby('short_label')[\n",
    "    ['percent_societal', 'percent_societal_06', 'percent_societal_08']\n",
    "].mean().reset_index()\n",
    "\n",
    "# Reshape to long format for grouped barplot\n",
    "means_long = means.melt(id_vars='short_label', \n",
    "                        value_vars=['percent_societal', 'percent_societal_06', 'percent_societal_08'],\n",
    "                        var_name='Threshold', \n",
    "                        value_name='Avg Societal Engagement')\n",
    "\n",
    "# Rename threshold labels for display\n",
    "threshold_map = {\n",
    "    'percent_societal': 'Raw',\n",
    "    'percent_societal_06': 'Threshold 0.6',\n",
    "    'percent_societal_08': 'Threshold 0.8'\n",
    "}\n",
    "means_long['Threshold'] = means_long['Threshold'].map(threshold_map)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7424e9e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== Average Societal Orientation by Team Type and Threshold ===\n",
      "Threshold         Raw  Threshold 0.6  Threshold 0.8\n",
      "short_label                                        \n",
      "CS-only          7.78           6.66           4.35\n",
      "Interd. (Full)  12.53          11.18           7.96\n",
      "Interd. (NSM)   13.18          11.61           7.97\n",
      "Interd. (SSH)   20.92          18.88          13.67\n"
     ]
    }
   ],
   "source": [
    "# Define color-blind-friendly palette for thresholds\n",
    "threshold_palette = {\n",
    "    'Raw': '#E69F00',          # orange\n",
    "    'Threshold 0.6': '#56B4E9',# sky blue\n",
    "    'Threshold 0.8': '#009E73' # bluish green\n",
    "}\n",
    "\n",
    "# Plot setup\n",
    "sns.set(style='whitegrid')\n",
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "ax = sns.barplot(\n",
    "    data=means_long,\n",
    "    x='short_label',\n",
    "    y='Avg Societal Engagement',\n",
    "    hue='Threshold',\n",
    "    palette=threshold_palette,\n",
    "    errorbar=None\n",
    ")\n",
    "\n",
    "# Style\n",
    "ax.set_title('Average Societal Orientation by Team Type and Threshold', fontsize=TITLE_SIZE)\n",
    "ax.set_xlabel('Team Type', fontsize=LABEL_SIZE)\n",
    "ax.set_ylabel('Average Percent', fontsize=LABEL_SIZE)\n",
    "ax.tick_params(axis='both', labelsize=TICK_SIZE)\n",
    "ax.legend(title='Threshold', fontsize=LEGEND_SIZE, title_fontsize=LEGEND_TITLE_SIZE)\n",
    "plt.xticks(rotation=20)\n",
    "plt.tight_layout()\n",
    "\n",
    "# Paths\n",
    "color_path = f\"{GRAPHICS_DIR}/avg_societal_orientation_groupedbars.png\"\n",
    "gray_path = f\"{GRAPHICS_DIR}/avg_societal_orientation_groupedbars_grayscale.png\"\n",
    "\n",
    "# Save\n",
    "plt.savefig(f\"{GRAPHICS_DIR}/avg_societal_orientation_groupedbars.pdf\", format='pdf', dpi=300, bbox_inches='tight')\n",
    "plt.savefig(color_path, format='png', dpi=300, bbox_inches='tight')\n",
    "\n",
    "# Convert to grayscale\n",
    "img = Image.open(color_path).convert(\"L\")\n",
    "img.save(gray_path)\n",
    "\n",
    "plt.show()\n",
    "\n",
    "# Print group-level means\n",
    "print(\"\\n=== Average Societal Orientation by Team Type and Threshold ===\")\n",
    "pivot_means = means_long.pivot(index='short_label', columns='Threshold', values='Avg Societal Engagement')\n",
    "print(pivot_means.round(2).to_string())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "4afd1b06",
   "metadata": {},
   "outputs": [],
   "source": [
    "##### Run the regression analysis with all three thresholds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "6ad8ee9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0.1', 'Unnamed: 0', 'abstract', 'author_ids',\n",
       "       'citation_count', 'conclusion', 'introduction', 'authors_final_FOS',\n",
       "       'paper_id', 'percent_societal', 'RQ_societal', 'RQ', 'subdomain',\n",
       "       'title', 'venue', 'year', 'team_size', 'event_type',\n",
       "       'percent_societal_new', 'total_sentences', 'article_length',\n",
       "       'categories', 'cs.cv', 'cs.cl', 'cs.cy', 'cs.ai', 'extracted_FOS',\n",
       "       'standardized_FOS', 'interdisciplinary_category', 'cs.cv', 'cs.ai',\n",
       "       'cs.cl', 'cs.cy', 'percent_societal_06', 'percent_societal_08'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "64aab387",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                          PanelOLS Estimation Summary                           \n",
      "================================================================================\n",
      "Dep. Variable:       percent_societal   R-squared:                        0.3757\n",
      "Estimator:                   PanelOLS   R-squared (Between):              0.1286\n",
      "No. Observations:               99448   R-squared (Within):               0.3680\n",
      "Date:                Fri, Jul 04 2025   R-squared (Overall):              0.3475\n",
      "Time:                        09:27:15   Log-likelihood                -3.956e+05\n",
      "Cov. Estimator:            Unadjusted                                           \n",
      "                                        F-statistic:                      6648.9\n",
      "Entities:                           2   P-value                           0.0000\n",
      "Avg Obs:                    4.972e+04   Distribution:                 F(9,99427)\n",
      "Min Obs:                       1920.0                                           \n",
      "Max Obs:                    9.753e+04   F-statistic (robust):             6648.9\n",
      "                                        P-value                           0.0000\n",
      "Time periods:                      11   Distribution:                 F(9,99427)\n",
      "Avg Obs:                       9040.7                                           \n",
      "Min Obs:                       2139.0                                           \n",
      "Max Obs:                     1.99e+04                                           \n",
      "                                                                                \n",
      "                                               Parameter Estimates                                                \n",
      "==================================================================================================================\n",
      "                                                Parameter  Std. Err.     T-stat    P-value    Lower CI    Upper CI\n",
      "------------------------------------------------------------------------------------------------------------------\n",
      "Fully Interdisciplinary                            5.0310     0.1912     26.308     0.0000      4.6562      5.4058\n",
      "Interdisciplinary (Natural Sciences/Medicine)      3.3258     0.1398     23.791     0.0000      3.0518      3.5998\n",
      "Interdisciplinary (Social Sciences/Humanities)     5.5917     0.2228     25.098     0.0000      5.1550      6.0284\n",
      "team_size                                          0.1233     0.0112     11.046     0.0000      0.1014      0.1451\n",
      "article_length                                    -0.0003  1.426e-05    -22.755     0.0000     -0.0004     -0.0003\n",
      "cs.cv                                             -3.4660     0.1390    -24.938     0.0000     -3.7385     -3.1936\n",
      "cs.cl                                             -3.8371     0.1527    -25.127     0.0000     -4.1364     -3.5378\n",
      "cs.ai                                              2.0174     0.1197     16.855     0.0000      1.7828      2.2520\n",
      "cs.cy                                              31.248     0.1811     172.52     0.0000      30.893      31.603\n",
      "==================================================================================================================\n",
      "\n",
      "F-test for Poolability: 148.99\n",
      "P-value: 0.0000\n",
      "Distribution: F(11,99427)\n",
      "\n",
      "Included effects: Entity, Time\n"
     ]
    }
   ],
   "source": [
    "data = merged_data.copy()\n",
    "data = data.loc[:, ~data.columns.duplicated()]\n",
    "# Ensure 'year' is an integer\n",
    "data = data.dropna(subset=['year'])\n",
    "data['year'] = data['year'].astype(int)\n",
    "\n",
    "# Set index for panel data (event_type as entity effect, year as time effect)\n",
    "data_ = data.set_index(['event_type', 'year'])\n",
    "\n",
    "# Dependent variable\n",
    "Y = data_['percent_societal']\n",
    "\n",
    "# Independent variables\n",
    "X = pd.get_dummies(data_['interdisciplinary_category'], drop_first=True)  # Interdisciplinary categories\n",
    "X['team_size'] = data_['team_size']  # Control variable: team size\n",
    "if 'article_length' in data_.columns:  # Add article length if available\n",
    "    X['article_length'] = data_['article_length']\n",
    "\n",
    "# Include fixed effects for categories ('cs.cv', 'cs.cl', 'cs.ai', 'cs.cy')\n",
    "for category in ['cs.cv', 'cs.cl', 'cs.ai', 'cs.cy']:\n",
    "    if category in data_.columns:\n",
    "        X[category] = data_[category]\n",
    "\n",
    "# Fit the PanelOLS model (with entity and time effects)\n",
    "model = PanelOLS(Y, X, entity_effects=True, time_effects=True).fit()\n",
    "print(model.summary)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "d2354f86",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                           PanelOLS Estimation Summary                           \n",
      "=================================================================================\n",
      "Dep. Variable:     percent_societal_06   R-squared:                        0.3705\n",
      "Estimator:                    PanelOLS   R-squared (Between):              0.1683\n",
      "No. Observations:                99448   R-squared (Within):               0.3628\n",
      "Date:                 Fri, Jul 04 2025   R-squared (Overall):              0.3537\n",
      "Time:                         09:27:54   Log-likelihood                -3.888e+05\n",
      "Cov. Estimator:             Unadjusted                                           \n",
      "                                         F-statistic:                      6502.7\n",
      "Entities:                            2   P-value                           0.0000\n",
      "Avg Obs:                     4.972e+04   Distribution:                 F(9,99427)\n",
      "Min Obs:                        1920.0                                           \n",
      "Max Obs:                     9.753e+04   F-statistic (robust):             6502.7\n",
      "                                         P-value                           0.0000\n",
      "Time periods:                       11   Distribution:                 F(9,99427)\n",
      "Avg Obs:                        9040.7                                           \n",
      "Min Obs:                        2139.0                                           \n",
      "Max Obs:                      1.99e+04                                           \n",
      "                                                                                 \n",
      "                                               Parameter Estimates                                                \n",
      "==================================================================================================================\n",
      "                                                Parameter  Std. Err.     T-stat    P-value    Lower CI    Upper CI\n",
      "------------------------------------------------------------------------------------------------------------------\n",
      "Fully Interdisciplinary                            4.6077     0.1787     25.785     0.0000      4.2574      4.9579\n",
      "Interdisciplinary (Natural Sciences/Medicine)      2.9311     0.1306     22.439     0.0000      2.6751      3.1872\n",
      "Interdisciplinary (Social Sciences/Humanities)     5.1106     0.2082     24.547     0.0000      4.7025      5.5186\n",
      "team_size                                          0.1016     0.0104     9.7415     0.0000      0.0811      0.1220\n",
      "article_length                                    -0.0003  1.332e-05    -20.822     0.0000     -0.0003     -0.0003\n",
      "cs.cv                                             -3.0468     0.1299    -23.458     0.0000     -3.3013     -2.7922\n",
      "cs.cl                                             -3.4094     0.1427    -23.892     0.0000     -3.6891     -3.1297\n",
      "cs.ai                                              1.8442     0.1118     16.488     0.0000      1.6249      2.0634\n",
      "cs.cy                                              29.091     0.1693     171.88     0.0000      28.759      29.423\n",
      "==================================================================================================================\n",
      "\n",
      "F-test for Poolability: 148.19\n",
      "P-value: 0.0000\n",
      "Distribution: F(11,99427)\n",
      "\n",
      "Included effects: Entity, Time\n"
     ]
    }
   ],
   "source": [
    "# Dependent variable\n",
    "Y = data_['percent_societal_06']\n",
    "\n",
    "# Independent variables\n",
    "X = pd.get_dummies(data_['interdisciplinary_category'], drop_first=True)  # Interdisciplinary categories\n",
    "X['team_size'] = data_['team_size']  # Control variable: team size\n",
    "if 'article_length' in data_.columns:  # Add article length if available\n",
    "    X['article_length'] = data_['article_length']\n",
    "\n",
    "# Include fixed effects for categories ('cs.cv', 'cs.cl', 'cs.ai', 'cs.cy')\n",
    "for category in ['cs.cv', 'cs.cl', 'cs.ai', 'cs.cy']:\n",
    "    if category in data_.columns:\n",
    "        X[category] = data_[category]\n",
    "\n",
    "# Fit the PanelOLS model (with entity and time effects)\n",
    "model = PanelOLS(Y, X, entity_effects=True, time_effects=True).fit()\n",
    "print(model.summary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "6b9bc296",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                           PanelOLS Estimation Summary                           \n",
      "=================================================================================\n",
      "Dep. Variable:     percent_societal_08   R-squared:                        0.3372\n",
      "Estimator:                    PanelOLS   R-squared (Between):              0.2890\n",
      "No. Observations:                99448   R-squared (Within):               0.3298\n",
      "Date:                 Fri, Jul 04 2025   R-squared (Overall):              0.3472\n",
      "Time:                         09:28:04   Log-likelihood                -3.696e+05\n",
      "Cov. Estimator:             Unadjusted                                           \n",
      "                                         F-statistic:                      5621.5\n",
      "Entities:                            2   P-value                           0.0000\n",
      "Avg Obs:                     4.972e+04   Distribution:                 F(9,99427)\n",
      "Min Obs:                        1920.0                                           \n",
      "Max Obs:                     9.753e+04   F-statistic (robust):             5621.5\n",
      "                                         P-value                           0.0000\n",
      "Time periods:                       11   Distribution:                 F(9,99427)\n",
      "Avg Obs:                        9040.7                                           \n",
      "Min Obs:                        2139.0                                           \n",
      "Max Obs:                      1.99e+04                                           \n",
      "                                                                                 \n",
      "                                               Parameter Estimates                                                \n",
      "==================================================================================================================\n",
      "                                                Parameter  Std. Err.     T-stat    P-value    Lower CI    Upper CI\n",
      "------------------------------------------------------------------------------------------------------------------\n",
      "Fully Interdisciplinary                            3.4545     0.1472     23.462     0.0000      3.1659      3.7430\n",
      "Interdisciplinary (Natural Sciences/Medicine)      1.9317     0.1076     17.948     0.0000      1.7208      2.1427\n",
      "Interdisciplinary (Social Sciences/Humanities)     3.6919     0.1715     21.522     0.0000      3.3557      4.0281\n",
      "team_size                                          0.0616     0.0086     7.1713     0.0000      0.0448      0.0785\n",
      "article_length                                    -0.0002  1.098e-05    -16.270     0.0000     -0.0002     -0.0002\n",
      "cs.cv                                             -1.9653     0.1070    -18.365     0.0000     -2.1750     -1.7555\n",
      "cs.cl                                             -2.3445     0.1176    -19.940     0.0000     -2.5750     -2.1141\n",
      "cs.ai                                              1.4352     0.0922     15.573     0.0000      1.2545      1.6158\n",
      "cs.cy                                              22.745     0.1395     163.10     0.0000      22.472      23.019\n",
      "==================================================================================================================\n",
      "\n",
      "F-test for Poolability: 140.24\n",
      "P-value: 0.0000\n",
      "Distribution: F(11,99427)\n",
      "\n",
      "Included effects: Entity, Time\n"
     ]
    }
   ],
   "source": [
    "# Dependent variable\n",
    "Y = data_['percent_societal_08']\n",
    "\n",
    "# Independent variables\n",
    "X = pd.get_dummies(data_['interdisciplinary_category'], drop_first=True)  # Interdisciplinary categories\n",
    "X['team_size'] = data_['team_size']  # Control variable: team size\n",
    "if 'article_length' in data_.columns:  # Add article length if available\n",
    "    X['article_length'] = data_['article_length']\n",
    "\n",
    "# Include fixed effects for categories ('cs.cv', 'cs.cl', 'cs.ai', 'cs.cy')\n",
    "for category in ['cs.cv', 'cs.cl', 'cs.ai', 'cs.cy']:\n",
    "    if category in data_.columns:\n",
    "        X[category] = data_[category]\n",
    "\n",
    "# Fit the PanelOLS model (with entity and time effects)\n",
    "model = PanelOLS(Y, X, entity_effects=True, time_effects=True).fit()\n",
    "print(model.summary)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "69ebbd51",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "societal_influence_env",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.21"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
